Antidepressants may alleviate depression by altering the function of the brain serotonin (5HT) or noradrenergic (NE) systems. The antidepressant response to specific 5-HT reuptake inhibitors such as fluoxetine (FLU) is temporarily reversed while that to the NE reuptake inhibitor, desipramine (DMI), is not, with rapid depletion of the precursor of 5-HT synthesis, tryptophan. As an extension of that data, we plan to use the drug alpha- methyl-para-tyrosine (AMPT), a reversible inhibitor of the rate limiting enzyme of catecholamine synthesis, tyrosine hydroxylase, to lower brain NE and dopamine (DA). The purpose of this study is to assess the importance of the NE and DA systems in the symptoms of depression and for the maintenance of the antidepressant response to DMI and FLU. We hypothesize that if the antidepressant response to DMl is dependent on changes induced in the brain NE system, then AMPT treatment should temporarily reverse that antidepressant response in a manner similar to that seen for FLU with tryptophan depletion. Methods: Sixty depressed patients (DSM-III-R criteria) will be tested with AMPT and control tests while drug-free and then again after successful antidepressant treatment with either DMI or FLU. After drug-free testing, patients are randomly assigned to either DMI or FLU and antidepressant treatment will be performed with open routine clinical dosing over a maximum of 10-weeks. Behavioral ratings of mood will be obtained weekly throughout the treatment phase of the study. Treatment response will be ascertained by predetermined response criteria. Treatment responders will be re-tested and antidepressant medications will be continued throughout the repeat tests. AMPT and control testing will be accomplished in a double-blind, crossover fashion. Each test will include a four day sequence with AMPT 1 gm TID (AMPT test) or diphenhydramine 50 mg TID (control test) administered on the middle two days and no AMPT or diphenhydramine on the first and last days (baseline and follow-up days). Diphenhydramine is used as an "active placebo" in order to maintain the blind given the sedation associated with AMPT. Behavioral ratings of mood and plasma for MHPG and HVA levels will be obtained at 9 AM and 4 PM during the middle two days and at 9 am on the first and last days. Expected Conclusions and Pilot Data: We expect that interference with the synthesis of NE and DA with AMPT will temporarily reverse antidepressant response to DMI but not FLU. Ten antidepressant-remitted patients (3 DMI, 3 FLU, 2 sertraline, 2 mazindol) and six drug-free depressed patients have been tested in a pilot study. Drug-free patients were unable to distinguish AMPT from diphenhydramine testing. The three patients having responded to DMI and the two mazindol-responders experienced an increase in depressive symptoms within 12 hours after starting AMPT and had a rapid recovery back to improved state within 12 hours of its discontinuation. The patients on FLU and sertraline became tired and sleepy but not depressed during both AMPT and diphenhydramine testing.